It's jarring to me that you so meticulously analyze the cost of dosing with melatonin; once I know a cost is below some low threshold, I prefer not to think about it at all.

That's great for you - if you feel convinced, you save all the time I spent researching & writing this. But how do you know the cost is below that threshold? Because I did so meticulously analyze it.

More to the point, I don't analyze everything like this. Once I calculated the cost per night was below 10 cents and the benefit above half an hour, my mind was made up. Everything after that was for a hopefully educational 'case study' for readers, and to make the conclusion as strong as possible.

I'd rather you took the same care into performing some objective tests of mental capability on varying amounts of sleep, so that it really means something when you say you gain an hour of wakefulness. Of course, I'd want this blinded as well, but I doubt you have convincing placebo pills available; besides, I don't mind taking something in hope of accruing some real and placebo benefits.

I do have visually convincing placebos (some vitamin B supplements), but part of the problem is there's no way I can fool myself - I know what the onset feels like with and without melatonin.

Although it wouldn't be a bad idea to take some measurements on the Dual N-back test; I would like to know the saving more precisely than 'more than half and hour and less than 2'. :)